Change of carbon mass after timber harvesting in a natural forest, West Sumatra Indonesia

Forests have an important role in the carbon cycle and in the dynamics of storing and releasing carbon in growth, decay, and disturbance processes. Timber harvesting initially reduces the amount of carbon in the forest. This study aims to analyze the magnitude of changes in carbon mass due to timber harvesting. It was conducted in a forest concessionaire located at Siberut of West Sumatra on a natural forest covering about 48,000 hectares. Selective cutting was implemented for logging where the harvestable trees were over 50 cm in diameter. The amount of carbon mass was obtained based on laboratory tests on tree parts, including trunk, branches, twigs, and bark. The average carbon content in each part of the tree is different. The trunk contains 53.86% of carbon mass, the branches are 51.98%, the twigs are 31.58%, the leaves are 27.91%, and the bark is 32.01%. Timber harvesting was conducted at the forest compartment with an average forest stand density of 71.5 trees ha−1 and a felling intensity of 8.8 trees ha−1, causing damage to 16.17 trees ha−1 stands. Timber harvesting causes a decrease in carbon mass reserves of 43.26% from forest carbon mass of 141.89 tons C ha−1 before harvesting to 80.00 tons C ha−1 after harvesting.


Introduction
Forest operation is a forest harvesting activity which is also known as logging.Timber harvesting in natural forests in Indonesia is still ongoing today.Harvesting natural forests extracts wood from the forest for use in the timber industry.Communities need timber from natural forests and plantation forests [1], and forest operations positively benefit society and the country.On the other hand, this forest harvesting impacts environmental damage [2].Other damage is in the form of damage to remaining stands [3,4], changes in river water discharge and quality [5], forest fires [6], harvesting waste [7], damage to forest soil in the form of soil compaction [8], changes in diversity [9], and other negative impacts.Timber harvesting causes a reduction in forest carbon mass because wood is extracted from the forest [10], and damage to remaining stands and opening of forest areas occurs [11].This timber harvesting causes carbon emissions to increase [12,13].
The increasing emission of Carbon Dioxide (CO2) into the atmosphere and its implications for the global climate has become an essential issue at the international level.Greenhouse gas (CO2) emissions from the forestry sector are estimated to be around 17% of global emissions [14].The important role of forests is not only as a carbon store but also as the most efficient carbon dioxide (CO2) absorber on earth and a source of greenhouse gases when not managed properly.The release of forest carbon into the atmosphere, or emission, occurs through various mechanisms such as the respiration of living things, decomposition of organic matter, and burning of biomass [15].Logging activities can cause damage to remaining stands and the death of trees.The dead tree will then release carbon into the atmosphere through decomposition.Timber harvesting causes a decrease in above-ground carbon stocks, which is 50% of the original carbon stock [16].
One currently developing effort to reduce carbon emissions is Reducing Emissions from Deforestation and Forest Degradation (REDD+).REDD+, as a general concept, encompasses various local, national, and global actions to reduce emissions caused by deforestation and forest degradation and increase forest carbon stocks in developing countries.REDD+ aims to offset greenhouse gas emissions by reducing emissions from deforestation and forest degradation [17].Efforts to estimate carbon for carbon trading using REDD+ need to be implemented with the best possible accuracy and precision but also need to consider costs [18].
Forest carbon measurement can be done using tree volume estimation methods, extra methods (laboratory tests) to estimate biomass, using volume tables, and using average tree weight [19].According to Brown, the assumption that carbon mass equals 50% of biomass in estimating carbon mass can lead to inaccurate results [20].Therefore, it is necessary to have a study on the measurement of carbon in natural forests through laboratory testing.

Study area
The study was performed in a natural forest in West Sumatra, Indonesia (Figure 1).It was conducted at the concessionaire forest, Mentawai Regency, a natural forest covering about 48,000 hectares.Selective cutting was implemented for logging where the harvestable trees were set at trees over 50 cm in diameter.Felling was carried out using a chainsaw, bucking, and end cutting.Skidding was performed using a bulldozer.

Field and laboratory measurement.
Sample plot was made on an annual logging block area which consists of 6 plots; each plot was 1 hectare (100m x 100 meters).Before the logging, all trees with a diameter of 20 cm dbh were calculated and measured within each sample plot.Each plot measured its stand density, logged tree species and calculated the log volume.After felling, total felled trees and residual stand damages were counted.The number of damaged and evicted trees along the skid trail was counted after skidding.The total number of damaged trees was the sum of the number of damaged trees during felling and skidding activities.Destructive methods carried out carbon measurements by cutting trees.In this study, the number of sample trees used was 30, with as many as 11 tree species that are the most dominant and follow the distribution of diameter classes.The number of trees is eligible for measuring carbon biomass and mass of 30−100 trees or at least 30 trees and must represent each type of tree.This method is used to measure natural vegetation or natural forests [15,19].Artocarpus integer Cempedak, Peiki 1 Total tree samples 30 In Table 1, it can be seen the dominant tree species in the study location.The most dominant tree is the type of keruing (Dipterocarpus elongatus Korth).Each of these sample trees consists of various diameter classes, so based on the data in Table 3, the sample trees can be grouped into five diameter classes.The division of diameter classes can be made with diameter intervals of every 10 cm [19,20].Data based on the distribution of diameter classes is presented in Table 2.
The number of sample trees in each diameter class is different because the composition of the tree diameters in the sample plots is different

Collection of the part logged tree for laboratory test
The parts of the logged tree that are measured as test samples in the laboratory include the stem, branches, twigs, leaves, and bark.Sampling for laboratory tests from each part of the tree was carried out in the following manner.The stem, branch, and twig samples were cut across to size 5x5x5 cm.The leaves were taken from mixed leaves as much as 300 grams.All samples were wrapped in aluminum foil and plastic.Carbon content measurement is carried out in the laboratory through several stages.Wood Specific Gravity measurements were carried out according to the American Society for Testing Materials (ASTM) D2395 [21], wood water content ASTM D4442 [22], volatile matter content ASTM D5832 [23], and ash content based on ASTM D2866 [24].Carbon content calculation of test samples from each part of the tree using the Indonesian National Standard (SNI) 06-3730-1995, the carbon content of the test sample results from a 100% reduction in the levels of flying substances and ash content.Carbon content can be calculated using the following formula: Carbon content (%) = 100% -levels of flying substances (%) -ash content (%) Carbon mass above the ground was calculated using the following formula: Carbon mass (ton) = biomass (ton) x carbon content (%) The log volume calculation using Brereton metric formula:

Stand damage
Stand damage is damage to the trees around the trees which were cut down and skidding activity.Differences in felling intensity cause different amounts of stand damage.Stand damage at the felling intensity (logged tress) of 8.8 ± 4.71 trees ha-1 is 24 ± 8.98 trees ha-1 (Table 3).The most damaged trees are trees in the diameter class due to the stand density of trees (Table 3).The linier regression showed the relationship between felling intensity (treesha -1 ) stand damage (treesha -1 ) (Figure 2).Relationship between felling intensity (treesha -1 ) stand damage (treesha -1 ).

Carbon content
Laboratory test results for the average carbon content of tree parts based on sample tree species are presented in Table 4. Felling intensity (trees ha -1 ) Laboratory test results for sample tree carbon content in each part of the tree based on diameter class are presented in Table 5.Table 5 shows that the average carbon content in each part of the tree differs; the highest carbon content is found in the stem, an average of 53.9 % ranging from 52.1% to 55.7%.This content is due to the low levels of volatile matter and ash in the stem.In addition, wood, especially the stem, comprises components such as lignin, cellulose, hemicellulose, and other extractive substances that contain lots of carbon.The main stem generally has more wood constituents than other parts of the tree (branches and twigs) [25].Leaves have the lowest carbon content, on average, 27.9%, with an average range from 27.0 to 29.3%.The photosynthetic products must be immediately distributed to other parts of the tree.The low lignin and cellulose content also cause low leaf carbon levels.There are differences in carbon content in differences in tree diameter classes.In each diameter class, there are differences in carbon content in various parts of the tree.In each diameter class, the largest carbon content is in the stem but the smallest in the leaves (Figure 3).This study's results align with the research by Yuniawati et al. [26], which stated that the highest carbon content was found in the stem, namely 50.47% and 60.20%.This carbon content is the same as the results of research by Kusuma [27] and Muhdi [28], which stated that the highest carbon content was found in the stems at 47.15% and 45.75%.The results of laboratory tests can show that the biomass content of tree parts and the carbon content of tree biomass in various diameter classes are different.Based on this, biomass's carbon content is important information in estimating the potential carbon mass of trees in stands.According to Yuniawati et al. [26], the assumption that is often used refers to [20] states that carbon mass is considered equal to 50% of biomass or has a conversion factor of 0.5 in estimating the potential carbon mass of a stand without regard to the type of biomass and age of the stand causing inaccurate results prediction.

Biomass, Carbon mass Before and After logging
Forest biomass is the total dry weight of all plant parts.Forest biomass is usually expressed in dry weight per unit area (ton ha -1 ).The above−ground biomass of various diameter classes of the tree in each plot is presented in Table 6.The average biomass before logging was 259 tons ha -1 .The larger the diameter class, the higher the biomass.The largest biomass is found in the diameter class more than 60 cm, 180.96 tons ha -1 , and the smallest biomass is found in the 20−29 cm diameter class, 6.99 tons ha -1 .The results of this study showed that the above-ground biomass before logging was lower than those of research by Muhdi [28] in Kalimantan and by Hertel [29] in Sulawesi, respectively 301.60 tons ha-1 and 303 tons ha-1.According to Mazzei et al. [30], above-ground biomass before logging in the Amazon rainforest, Brazil, of 409.8 tons ha-1 of timber harvesting causes a loss of biomass of 69.3 tons ha-1 due to timber harvesting.Biomass stands before logging in Mont's de Cristal, Gabon, according to Medjibe et al. [10], an average of 420.4 tons ha-1.This study's results indicate lower biomass than other studies because this study only calculates above-ground biomass for all trees with a diameter greater than 20 cm, excluding shrubs and grasses.Carbon mass before logging is the total carbon of trees per unit area.The results of total carbon mass based on diameter class in each plot are presented in Table 7. Table 7 shows that each plot's average total carbon mass is 141.89 tons ha-1, ranging from 105.22−204.79tons C ha-1.This carbon mass is the carbon mass above the ground before logging.
The results of this study are greater than those of Hertel et al. [29], which state that the average carbon mass in Sulawesi forests is 120.0 tons C ha-1.However, this study aligns with Lasco's statement [16] that the carbon mass of tropical forests in Asia ranges from 40−250 tons C ha-1.

Carbon mass after logging
Logging activities in natural forests cause damage to both the residual stands, impacting changes in carbon stocks, soil, and biodiversity in the harvested area.The carbon mass of trees damaged by felling and skidding is presented in Table 8.The carbon mass of the stand damaged by tree felling is 8.85 tons C ha -1 , while the carbon mass of the residual stand damaged by skidding is 7.32 tons C ha -1 .The total mass of carbon due to damage by logging activities is 16.17 tons C ha -1 (Table 8).The results of this study show lower results when compared to research [26], namely the potential carbon emissions of 16.86 ton C ha -1 .

Carbon mass of logged trees
Logging activities cause carbon stocks stored in living vegetation to decrease and result in an increase in dead trees.Logging can be done to the Indonesian selective cutting rules.The forest concessionaire only logs commercial trees larger than 50 cm in diameter.The carbon mass of trees cut in each study plot is presented in Table 9. Logged trees cause the reserves of carbon stock in the forest to be The carbon mass of logged trees is 45.78ton C ha -1 in the 21.98− 68.22ton C ha -1 range.This study shows higher results when compared to research [26,27,29], namely the potential carbon emissions resulting from trees being cut down at the same location.

Carbon Mass Change
The logged trees contributed 45.78 tons C ha-1 and stand damage 16.17 tons C ha-1 (Figure 4).The total loss of carbon mass was 61.95 tons C ha-1.This research showed that carbon mass decreased by 43.26%.These results are lower compared to other research [30,31,32,33], which measured carbon mass in all vegetation, litter and necro mass, and waste; the carbon mass reduction was 54.79%, that is from carbon mass of 135.87 tons C ha-1 before harvesting to 61.42 tons C ha-1.The difference results because of only measured the above-ground biomass and not all vegetation but only above 20 cm.One of the logging techniques that Reduce Impact logging needs to be applied in order to reduce damage to the stands [34,35,36] and reduce the loss of biomass or carbon mass due to logging activities [37,38,39,40].

Conclusions
The stem of the tree has the highest carbon content.It was 54.03% of the carbon mass of the stem, and the leaves had 27.33%, the lowest carbon content.Biomass and carbon mass before harvesting are 259.64 tons C ha -1 and 141.89 tons C ha -1 .Logging causes a decrease in carbon mass reserves of 43.26%, from forest carbon mass of 141.89 tons C ha -1 before logging to 79.94 tons C ha -1 or about 80.00 tons C ha -1 after logging.Carbon mass (ton C/ha)
log (m 3 ) LBp = Large end area of the log (m 2 ) LBu = Small end area of the log (m 2 ) p = Length of log (m) Dp = Average large end diameter (cm) Du = Average small end diameter (cm)

Figure 3 .
Figure 3. Percentage of Carbon content in each part of the tree at a natural forest of Siberut Island, Mentawai Regency, West Sumatera.

Figure 4 .
Figure 4. Carbon mass change before and after logging. before

Table 1 .
Tree species and number of sample trees.

Table 2 .
Number of sample trees in each diameter class.

Table 3 .
Felling intensity and stand damage by felling and skidding.

Table 4 .
Carbon content (%) of tree species in each part of trees.

Table 5
Carbon content (%) of diameter class in each part of trees.

Table 6 .
Above ground biomass of trees in each diameter class before logging*).
*) This above-ground biomass is only the biomass of a tree diameter of more than 20 cm.Exclude the biomass of shrubs and grass.

Table 7 .
Carbon mass of above-ground biomass of the tree in each class diameter before logging.

Table 8 .
Carbon mass of stand damage caused by logging operation

Table 9
Carbon mass of logged tree in each diameter class.